Method and device for constructing mechanism evaluation model

Abstract

The invention discloses a construction method and device of a mechanism evaluation model. The method comprises the following steps: acquiring sample mechanism basic elements, sample scientific research capability elements and sample mechanism bearing capacity elements of a sample mechanism; extracting a sample structured data set and a sample unstructured data set; converting the sample unstructured data set into a sample floating-point number vector, and converting the sample structured data set into a sample feature vector; inputting the sample floating-point number vector and the sample evaluation score label into an unstructured data evaluation model for score prediction, and inputting the sample feature vector and the sample evaluation score label into a structured data evaluation model for score prediction; a target weight ratio is screened out; and constructing a mechanism evaluation model according to the unstructured data evaluation model, the structured data evaluation model and the target weight ratio. According to the invention, the accuracy of the mechanism evaluation model is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanism evaluation, and in particular to a method and device for constructing a mechanism evaluation model. Background Art

[0002] Traditional research institution evaluation systems typically use a single metric metadata to assess research institutions. For example, for university-type research institutions, evaluation results are derived based on the total number of representative academic papers published and the total number of citations received. Traditional research institution evaluation systems that use a single metric metadata to derive evaluation results consider a relatively narrow range of factors and are unable to provide a reasonable and reliable evaluation of research institutions. Summary of the Invention

[0003] The present invention provides a method and device for constructing an institution evaluation model, which can improve the accuracy of the institution evaluation model, thereby enabling rapid and accurate prediction of the evaluation score of the institution to be evaluated.

[0004] In one aspect, the present invention provides a method for constructing an institution evaluation model, the method comprising:

[0005] Obtaining sample institution basic elements, sample scientific research capability elements, and sample institution carrying capacity elements of the sample institution as sample source data; the sample source data is annotated with a sample evaluation score label of the sample institution;

[0006] Extracting sample numerical data and sample character data as a sample structured data set based on the sample organization basic elements and sample scientific research capability elements;

[0007] Extracting sample long texts from the sample scientific research capability elements and the sample institutional carrying capacity elements as sample unstructured data sets;

[0008] Converting the sample unstructured data set into a sample floating-point vector, and converting the sample structured data set into a sample feature vector;

[0009] Inputting the sample floating-point number vector and the sample evaluation score label into the unstructured data evaluation model for score prediction to obtain a first sample score; and inputting the sample feature vector and the sample evaluation score label into the structured data evaluation model for score prediction to obtain a second sample score;

[0010] Traversing the combined weight set of the model, and screening a target weight ratio from multiple weight ratios according to each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label;

[0011] An institutional evaluation model is constructed based on the unstructured data evaluation model, the structured data evaluation model, and the target weight ratio.

[0012] Another aspect provides a device for constructing an organization evaluation model, the device comprising:

[0013] A sample data acquisition module is used to acquire sample institution basic elements, sample scientific research capability elements, and sample institution carrying capacity elements of a sample institution as sample source data; the sample source data is annotated with a sample evaluation score label of the sample institution;

[0014] A sample structured data determination module is used to extract sample numerical data and sample character data as a sample structured data set based on the sample organization basic elements and the sample scientific research capability elements;

[0015] A sample unstructured data determination module is used to extract the sample long text from the sample scientific research capability element and the sample institutional carrying capacity element as a sample unstructured data set;

[0016] A sample vector conversion module, configured to convert the sample unstructured data set into a sample floating-point vector, and convert the sample structured data set into a sample feature vector;

[0017] A sample score prediction module is configured to input the sample floating-point number vector and the sample evaluation score label into an unstructured data evaluation model for score prediction to obtain a first sample score; and input the sample feature vector and the sample evaluation score label into a structured data evaluation model for score prediction to obtain a second sample score;

[0018] a target weight ratio determination module, configured to traverse the combined weight set of the model and select a target weight ratio from a plurality of weight ratios based on each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label;

[0019] An evaluation model construction module is used to construct an institutional evaluation model based on the unstructured data evaluation model, the structured data evaluation model and the target weight ratio.

[0020] On the other hand, an electronic device is provided, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the method for constructing the institutional evaluation model as described above.

[0021] On the other hand, a computer storage medium is provided, which stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the method for constructing an institutional evaluation model as described above.

[0022] On the other hand, a computer program product is provided, comprising a computer program, wherein the computer program is loaded and executed by a processor to implement the method for constructing an institution evaluation model as described above.

[0023] The method and device for constructing an institutional evaluation model provided by the present invention have the following technical effects:

[0024] The present invention obtains the sample organization basic elements, sample scientific research capability elements and sample organization carrying capacity elements of the sample organization as sample source data; the sample source data is marked with the sample evaluation score label of the sample organization; according to the sample organization basic elements and sample scientific research capability elements, sample numerical data and sample character data are extracted as sample structured data sets; the sample long texts in the sample scientific research capability elements and sample organization carrying capacity elements are extracted as sample unstructured data sets; thereby dividing the sample source data into two major categories of data; converting the sample unstructured data set into a sample floating-point vector, and converting the sample structured data set into a sample feature vector; inputting the sample floating-point vector and the sample evaluation score label into an unstructured data evaluation model Perform score prediction to obtain a first sample score; and input the sample feature vector and the sample evaluation score label into the structured data evaluation model for score prediction to obtain a second sample score; thereby respectively obtaining the evaluation scores of the sample institutions obtained by the two types of data prediction; then traverse the combined weight set of the model, and according to each weight ratio in the combined weight set, the first sample score, the second sample score and the sample evaluation score label, screen out a target weight ratio from multiple weight ratios; construct an institution evaluation model according to the unstructured data evaluation model, the structured data evaluation model and the target weight ratio; the institution evaluation model constructed using the method of the present invention has a higher accuracy rate, thereby being able to quickly and accurately predict the evaluation score of the institution to be evaluated. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions and advantages of the embodiments of this specification or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0026] Figure 1This is a schematic diagram of an application environment of a method for constructing an institution evaluation model provided in an embodiment of this specification;

[0027] Figure 2 This is a flow chart of a method for constructing an institution evaluation model provided in an embodiment of this specification;

[0028] Figure 3 This is a flowchart of a method for training an unstructured data evaluation model provided in an embodiment of this specification;

[0029] Figure 4 This is a flowchart of a method for training a structured data evaluation model provided in an embodiment of this specification;

[0030] Figure 5 This is a flowchart of a method for screening a target weight ratio from multiple weight ratios based on each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label, provided by an embodiment of this specification;

[0031] Figure 6 This is a flow chart of a method provided by an embodiment of this specification for determining a combined evaluation index value of a model corresponding to each weight ratio according to each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label;

[0032] Figure 7 This is a flowchart of a method for determining an evaluation score of an organization to be evaluated provided in an embodiment of this specification;

[0033] Figure 8 This is a schematic diagram of a system for constructing an institution evaluation model provided in an embodiment of this specification;

[0034] Figure 9 This is a schematic diagram of a method for calculating a first sample score provided in an embodiment of this specification;

[0035] Figure 10 This is a schematic diagram of a device for constructing an organization evaluation model provided in an embodiment of this specification;

[0036] Figure 11 This is a structural diagram of a server provided in an embodiment of this specification. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0038] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way are interchangeable where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0039] The following describes a method for constructing an institutional evaluation model of the present invention. Figure 1 It is a flow chart of a method for constructing an institutional evaluation model provided in an embodiment of this specification. This specification provides method operation steps as described in the embodiment or flow chart, but may include more or fewer operation steps based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the steps among many orders, and does not represent the only order of execution. When the actual system or server product is executed, it can be executed sequentially or in parallel (for example, in a parallel processor or multi-threaded processing environment) according to the method shown in the embodiment or the accompanying drawings. Specifically, Figure 1 As shown, the method may include:

[0040] S101: Obtain sample institution basic elements, sample scientific research capability elements, and sample institution carrying capacity elements of a sample institution as sample source data; the sample source data is marked with a sample evaluation score label of the sample institution.

[0041] In an embodiment of this specification, the method of this embodiment can be applied to a system for constructing an institutional evaluation model. The system for constructing an institutional evaluation model of this embodiment may include an unstructured data evaluation model, a structured data evaluation model, and a weight calculation layer. The sample institution may be a scientific research institution, which may include but is not limited to scientific research departments of colleges and universities, independent scientific research institutes, and R&D departments within enterprises. The sample source data of the sample institution may include at least one of the sample institution basic elements, sample scientific research capability elements, and sample institution carrying capacity elements. Exemplarily, the sample source data is multi-source data, including sample institution basic elements, sample scientific research capability elements, and sample institution carrying capacity elements. The sample source data may include one or more institutional element library sources.

[0042] Among them, the basic elements of the sample institutions are set at two levels, including project source categories and basic information of the institutions. The project source categories are classified into the number of local tasks, the number of national tasks, the number of domestic commissioned projects, the number of independent deployments of the institute, the number of tasks of the institute units, and the number of other tasks; the basic information of the institutions is set at three levels, namely, the institutional team, the institutional budget, and the number of projects. The institutional team is classified into the number of senior professors, the number of deputy senior professors, the number of intermediate professors, the number of junior professors, the number of ordinary people, the number of doctors, the number of masters, and the number of undergraduates.

[0043] The scientific research capability elements of the sample institutions are set at five levels, including invention patents, published papers, monographs, scientific and technological reports, and research achievement transformation. Among them, invention patents include the number of domestic patent authorizations, the number of foreign patent authorizations, the number of invention patents, the number of utility model patents, the number of design patents, and the total number of citations; published papers include the total number of published papers and the total number of citations; monographs include the number of monographs; scientific and technological reports include the number of scientific and technological reports and the content of key core technology breakthroughs. Among them, the content of key core technology breakthroughs and research achievement transformation will be described in long text form.

[0044] The sample organization's carrying capacity elements are set at five levels, including project level, material resources, information resources, technical resources, and human resources. Material resources, information resources, technical resources, and human resources will be described in long text form. The project level is categorized into projects, subprojects, and tasks.

[0045] S102: Extracting sample numerical data and sample character data as a sample structured data set based on the sample organization basic elements and the sample scientific research capability elements.

[0046] In the embodiment of this specification, sample numerical data and sample character data can be extracted based on the sample organization basic element and the sample scientific research capability element. The sample character data and the sample numerical data can be a set of data with an associated relationship. For example, if the organization includes 30 undergraduates, and the sample character data is the number of undergraduates, then the sample numerical data is 30.

[0047] S103: Extracting sample long texts from the sample scientific research capability element and the sample institutional carrying capacity element as a sample unstructured data set.

[0048] In the embodiments of this specification, the sample long text is a text whose length is greater than a preset threshold; the sample scientific research capability element and the sample institutional carrying capacity element both include long text content, and the sample long text such as the breakthrough content of key core technologies and the transformation content of results in the sample scientific research capability element can be extracted as a sample unstructured data set; the material resources, information resources, technical resources, human resources, etc. in the sample institutional carrying capacity element are extracted as a sample unstructured data set.

[0049] S104: Convert the sample unstructured data set into a sample floating-point vector, and convert the sample structured data set into a sample feature vector.

[0050] In an embodiment of the present specification, a sample unstructured data set is converted into a sample floating-point number vector for input into an unstructured data evaluation model, and the sample structured data set is converted into a sample feature vector for input into a structured data evaluation model.

[0051] S105: Input the sample floating-point number vector and the sample evaluation score label into the unstructured data evaluation model for score prediction to obtain a first sample score; and input the sample feature vector and the sample evaluation score label into the structured data evaluation model for score prediction to obtain a second sample score.

[0052] In the embodiments of this specification, the unstructured data evaluation model is a model trained based on the unstructured data and historical sample evaluation score labels of the historical sample organization, and the structured data evaluation model is a model trained based on the structured data and historical sample evaluation score labels of the historical sample organization. The sample organization evaluation score, i.e., the first sample score, can be obtained by using the sample floating-point vector and the unstructured data evaluation model; and the sample organization evaluation score, i.e., the second sample score, can be obtained by using the sample feature vector and the structured data evaluation model. This facilitates subsequent adjustment of the weights of the two models based on the first sample score and the second sample score.

[0053] S106: Traverse the combined weight set of the model, and filter out a target weight ratio from multiple weight ratios according to each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label.

[0054] In an embodiment of the present specification, after obtaining the evaluation scores corresponding to the two models, the combined weight set can be input into the weight calculation layer to adjust the weights of the two models to obtain the optimal target weight ratio. The combined weight set of the model is a set of multiple preset weight ratios set for the unstructured data evaluation model and the structured data evaluation model. The weights corresponding to the unstructured data evaluation model and the structured data evaluation model can be determined based on each preset weight ratio in the combined weight set, where the sum of the weights of the two models is 1. The optimal target weight ratio can be screened out from multiple weight ratios based on each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label.

[0055] S107: Constructing an institutional evaluation model based on the unstructured data evaluation model, the structured data evaluation model, and the target weight ratio.

[0056] In an embodiment of the present specification, the first weight of the prediction result of the unstructured data evaluation model and the second weight of the preset result of the structured data evaluation model can be determined based on the target weight ratio corresponding to the unstructured data evaluation model and the structured data evaluation model, thereby constructing an institutional evaluation model for predicting institutional evaluation scores.

[0057] In some embodiments, extracting sample numerical data and sample character data as a sample structured data set based on the sample organization basic elements and the sample scientific research capability elements includes:

[0058] Extracting the number of projects, the number of tasks for each project, the budget for each project, and the number of people at each level in the organization team from the basic elements of the sample organization to obtain first sample structured data;

[0059] Extracting the number of scientific research achievements in the sample scientific research capability element to obtain second sample structured data;

[0060] The sample structured data set is constructed according to the first sample structured data and the second sample structured data.

[0061] In an embodiment of the present specification, the number of projects, the number of tasks for each project, the budget for each project, and the number of people at each level in the institutional team in the sample institutional basic element can be extracted to obtain first sample structured data; illustratively, the first sample structured data includes the number of local tasks, national tasks, domestic commissioned projects, research institutes' independent deployments, institute unit tasks, and other tasks corresponding to the project source categories in the sample institutional basic element; and the institutional team is classified into the number of senior staff, deputy senior staff, intermediate staff, junior staff, ordinary staff, doctoral staff, master's staff, and undergraduate staff;

[0062] The number of scientific research results in the sample scientific research capability element can be extracted to obtain the second sample structured data; illustratively, the second sample structured data includes the number of domestic patent authorizations, the number of foreign patent authorizations, the number of invention patents, the number of utility model patents, the number of design patents and the total number of citations included in the invention patents in the sample scientific research capability element; the total number of published papers and the total number of citations included in the published papers; the number of monographs included in the monographs; the number of scientific and technological reports included in the scientific and technological reports, etc.

[0063] Finally, a set formed by the first sample structured data and the second sample structured data is determined as the sample structured data set, thereby achieving accurate extraction of the sample structured data set from the sample source data.

[0064] In some embodiments, as Figure 2 As shown, the converting of the sample unstructured data set into a sample floating-point vector and the converting of the sample structured data set into a sample feature vector includes:

[0065] S1041: Inputting the sample unstructured data set into the unstructured data processing layer for data cleaning to obtain a first sample standard data set;

[0066] S1042: Inputting the first sample standard data set into a first vector representation layer to perform text semantic extraction to obtain the sample floating-point number vector;

[0067] S1043: Inputting the sample structured data set into the structured data processing layer for data cleaning to obtain a second sample standard data set;

[0068] S1044: Input the second sample standard data set into the second vector representation layer to perform data correlation analysis to obtain the sample feature vector.

[0069] In the embodiments of the present specification, the standard data set composed of the first sample standard data set and the second sample standard data set can also be a series of data preprocessing steps, including text cleaning, word segmentation, stop word removal, normalization, and missing value filling, etc., to convert the source data into a standardized data set that can be used by the model. The construction system of the institutional evaluation model of this embodiment also includes a data processing layer, which may include an unstructured data processing layer and a structured data processing layer; the data processing layer mainly performs data extraction and data preprocessing on the collected institutional data to output a data set that meets the model input requirements, laying the foundation for the subsequent construction of the evaluation model. The data preprocessing process performs different preprocessing operations based on the differences in data categories. For unstructured data, text segmentation, stop word removal, stem extraction, etc. are performed; for structured data, data cleaning, missing value filling, outlier processing, etc. are performed.

[0070] In the process of converting the sample unstructured data set into a sample floating-point vector, the sample unstructured data set can be first input into the unstructured data processing layer for data cleaning to obtain a first sample standard data set; then the first sample standard data set is input into the first vector representation layer for text semantic extraction to obtain the sample floating-point vector (high-dimensional vector); wherein the high-dimensional vector is obtained by processing long text input through a multilingual model, mapping sentences and paragraphs to a dense vector space, and thus obtaining a floating-point vector representation of the text. The unstructured data processing layer can be a pre-trained structural layer for cleaning the unstructured data set; the first vector representation layer can be a pre-trained structural layer for extracting text semantic features from the first sample standard data set, and the extracted text semantic features are sample floating-point vectors; the sample floating-point vectors are vectors that conform to the input data rules of the unstructured data evaluation model; thereby facilitating the unstructured data evaluation model to further predict the institutional evaluation score based on the sample floating-point vectors.

[0071] In the process of converting the sample structured data set into a sample feature vector, the sample structured data set can be first input into the structured data processing layer for data cleaning to obtain a second sample standard data set; then the second sample standard data set is input into the second vector representation layer for data correlation analysis to obtain a sample feature vector (feature vector) for characterizing data correlation; wherein the feature vector is extracted and converted by analyzing the inherent correlation of structured data. The structured data processing layer can be a pre-trained structural layer for cleaning the structured data set; the second vector representation layer can be a pre-trained structural layer for extracting correlation features from the second sample standard data set, and the extracted correlation features are sample feature vectors; the sample feature vector is a vector that conforms to the input data rules of the structured data evaluation model; thereby facilitating the structured data evaluation model to further predict the institution evaluation score based on the sample feature vector.

[0072] In some embodiments, as Figure 3 As shown, the training method of the unstructured data evaluation model includes:

[0073] S301: Inputting the sample floating-point number vector into an unstructured data evaluation network to perform evaluation score prediction to obtain a first prediction score;

[0074] S302: Determine first loss data based on a difference between the first prediction score and the sample evaluation score label;

[0075] S303: Adjust the parameters of the unstructured data processing layer, the first vector representation layer, and the unstructured data evaluation network according to the first loss data until the training end condition is met, and determine the unstructured data evaluation network at the end of training as the unstructured data evaluation model.

[0076] In an embodiment of the present specification, the unstructured data evaluation model is a model generated by training an unstructured data evaluation network based on a sample floating-point number vector and a sample target vector corresponding to a sample evaluation score label. The unstructured data evaluation network can be a deep learning algorithm network. Specifically, the sample floating-point number vector and the sample evaluation score label can be input into the unstructured data evaluation network to obtain a first vector corresponding to a first prediction score and a sample target vector corresponding to the sample evaluation score label, respectively. Then, the difference between the first vector and the sample target vector is calculated as the difference between the first prediction score and the sample evaluation score label to obtain the first loss data; then, according to the first loss data, the parameters of the unstructured data processing layer, the first vector representation layer and the unstructured data evaluation network are adjusted until the training end condition is met, and the unstructured data evaluation network at the end of training is determined as the unstructured data evaluation model. Wherein, the training end condition can be determined according to the first loss data and / or the number of training iterations.

[0077] In some embodiments, the unstructured data processing layer at the end of training, the first vector representation layer, and the unstructured data evaluation network can also be combined into a first score prediction model; during the application process, it is only necessary to input the unstructured data set of the organization to be evaluated into the first score prediction model to obtain the first evaluation score.

[0078] In some embodiments, as Figure 4 As shown, the training method of the structured data evaluation model includes:

[0079] S401: Inputting the sample feature vector into a structured data evaluation network to predict an evaluation score to obtain a second prediction score;

[0080] S402: Determine second loss data based on a difference between the second prediction score and the sample evaluation score label;

[0081] S403: Adjust the parameters of the structured data processing layer, the second vector representation layer, and the structured data evaluation network according to the second loss data until the training end condition is met, and determine the structured data evaluation network at the end of training as the structured data evaluation model.

[0082] In an embodiment of the present specification, the structured data evaluation model is a model generated by training a structured data evaluation network based on a sample target vector corresponding to a sample feature vector and a sample evaluation score label. The structured data evaluation network can be a linear model. Specifically, the sample feature vector and the sample evaluation score label can be input into the structured data evaluation network to obtain a second vector corresponding to the second prediction score and a sample target vector corresponding to the sample evaluation score label, respectively. The difference between the second vector and the sample target vector is then calculated as the difference between the second prediction score and the sample evaluation score label to obtain the second loss data; the parameters of the structured data processing layer, the second vector representation layer, and the structured data evaluation network are adjusted according to the second loss data until the training end condition is met, and the structured data evaluation network at the end of the training is determined as the structured data evaluation model. The training end condition can be determined based on the second loss data and / or the number of training iterations.

[0083] In some embodiments, the structured data processing layer at the end of training, the second vector representation layer, and the structured data evaluation network can also be combined into a second score prediction model; during the application process, it is only necessary to input the structured data set of the organization to be evaluated into the second score prediction model to obtain the second evaluation score.

[0084] In some embodiments, as Figure 5 As shown, the combined weight set of the ergodic model is filtered out from multiple weight ratios according to each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label, including:

[0085] S1061: traverse the combined weight set of the model, and determine the combined evaluation index value of the model corresponding to each weight ratio according to each weight ratio, the first sample score, the second sample score, and the sample evaluation score label in the combined weight set;

[0086] S1062: Determine the optimal weight ratio of the combined evaluation index value that meets the preset conditions as the target weight ratio.

[0087] In an embodiment of the present specification, the combined evaluation index value can be a numerical value corresponding to an evaluation index, or it can be an average or weighted average of multiple evaluation indexes; wherein, different institutions can choose different evaluation indexes according to actual conditions; the evaluation indexes may include but are not limited to indicators such as accuracy and recall rate; specifically, the combined weight set can be input into the weight calculation layer, and then the combined weight set of the model can be traversed, and the final sample evaluation score can be determined based on each weight ratio, the first sample score, and the second sample score in the combined weight set, and then the target loss data can be determined based on the difference between the sample evaluation score and the sample evaluation score label, and then the model parameters of the unstructured data evaluation model and the structured data evaluation model can be fine-tuned based on the target loss data; exemplarily, if the combined evaluation index value is accuracy, the end condition of the model fine-tuning training can be determined as the target loss data being less than a preset threshold, and the weight ratio at the end of the fine-tuning training can be determined as the optimal target weight ratio, and the unstructured data evaluation model and the structured data evaluation model at the end of the fine-tuning training can be used as the model for application.

[0088] In some embodiments, as Figure 6 As shown, traversing the combined weight set of the model, and determining the combined evaluation index value of the model corresponding to each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label, includes:

[0089] S10611: Inputting the combined weight set of the model into the weight calculation layer, traversing the combined weight set, and adjusting the weight ratio of the unstructured data evaluation model and the structured data evaluation model;

[0090] S10612: Determine, based on the first sample score, the second sample score, and each weight ratio, a sample prediction evaluation score for the sample institution at each weight ratio;

[0091] S10613: Determine the combined evaluation index value of the model corresponding to each weight ratio based on the difference between the sample prediction evaluation score under each weight ratio and the sample evaluation score label.

[0092] In an embodiment of the present specification, in the process of determining the target weight ratio, it is necessary to calculate the product of the first sample score and the first weight ratio to obtain the first product, and calculate the product of the second sample score and the second weight ratio to obtain the second product; then calculate the sum of the first product and the second product to obtain the sample prediction evaluation score; finally, the combined evaluation index value of the model corresponding to each weight ratio can be determined based on the difference between the sample prediction evaluation score under each weight ratio and the sample evaluation score label.

[0093] In some embodiments, after constructing the organization evaluation model based on the unstructured data evaluation model, the structured data evaluation model and the target weight ratio, Figure 7 As shown, the method includes:

[0094] S701: Obtain the institutional basic elements, scientific research capability elements, and institutional carrying capacity elements of the institution to be evaluated as source data to be evaluated;

[0095] S702: Extracting a structured data set based on the institutional basic elements and the scientific research capability elements; and extracting the long text in the scientific research capability elements and the institutional carrying capacity elements as an unstructured data set;

[0096] S703: Convert the unstructured data set into a floating-point vector, and convert the structured data set into a feature vector;

[0097] S704: Inputting the floating-point number vector into the unstructured data evaluation model to perform score prediction to obtain a first evaluation score;

[0098] S705: Inputting the feature vector into the structured data evaluation model to perform score prediction to obtain a second evaluation score;

[0099] S706: Obtaining the evaluation score of the organization to be evaluated according to the first evaluation score, the second evaluation score and the target weight ratio.

[0100] In the embodiments of this specification, the classification and preprocessing methods of the input data during the application of the model are the same as those of the sample source data. The first evaluation score of the organization to be evaluated can be predicted based on the unstructured data set, and the second evaluation score of the organization to be evaluated can be predicted based on the structured data set; and then the evaluation score of the organization to be evaluated can be obtained based on the first evaluation score, the second evaluation score and the target weight ratio; thus, the evaluation score of the organization to be evaluated can be quickly and accurately predicted through the two models.

[0101] In some embodiments, obtaining the evaluation score of the organization to be evaluated based on the first evaluation score, the second evaluation score, and the target weight ratio includes:

[0102] Determining a first weight of the unstructured data evaluation model and a second weight of the structured data evaluation model according to the target weight ratio;

[0103] Calculating a first product of the first evaluation score and the first weight, and a second product of the second evaluation score and the second weight;

[0104] The sum of the first product and the second product is calculated to obtain an evaluation score of the institution to be evaluated.

[0105] In an embodiment of the present specification, after determining the optimal target weight ratio of the two models, the first product of the first evaluation score and the first weight, and the second product of the second evaluation score and the second weight can be calculated; then the sum of the first product and the second product is calculated to obtain the evaluation score of the organization to be evaluated, thereby further improving the evaluation score of the organization to be evaluated through the weighted sum score calculation method.

[0106] Specifically, in the embodiments of this specification, Figure 8 As shown, Figure 8 This is a schematic diagram of the structure of a system for constructing an institutional evaluation model according to this embodiment. The system comprises an input layer, a data processing layer, a vector representation layer, a model layer, a weight calculation layer, and an output layer. The input layer is used to input long text data, character data, and numeric data. The data processing layer is used to preprocess the long text, numeric data, and character data in the source data to output a standard data set that meets the input criteria of the model layer. The vector representation layer is used to extract the semantic content of long text in unstructured data sets and convert it into high-dimensional vectors; analyze the relevance of structured data sets and convert it into feature vectors; and convert historical institutional ratings into target vectors. The model layer receives the high-dimensional vectors and feature vectors output by the vector representation layer, performs model training using deep learning and machine learning algorithms, and constructs unstructured data evaluation models and structured data evaluation models. The generated models are also evaluated for accuracy to ensure their effectiveness and reliability. The weight calculation layer is used to calculate the weight of each model based on its importance in the evaluation model. The output layer is used to output the final evaluation score.

[0107] The data processing layer primarily extracts and preprocesses collected institutional data to output a dataset that meets model input requirements, laying the foundation for subsequent evaluation model construction. The data preprocessing process performs different preprocessing operations depending on the data type. For unstructured data, this involves text segmentation, stop word removal, and stemming. For structured data, data cleaning, missing value filling, and outlier processing are performed.

[0108] The vector representation layer converts data into vectors suitable for AI algorithms. The standard dataset includes floating-point vectors, feature vectors, and target vectors.

[0109] For example, based on the sample institution basic elements, sample scientific research capability elements and sample institution carrying capacity elements, the sample structured data set and the sample unstructured data set can also be extracted and processed using a classification model; the classification model of the sample source data can be pre-trained, and then multiple sample source data such as the sample institution basic elements, sample scientific research capability elements and sample institution carrying capacity elements are input into the model to automatically divide the sample source data into the sample structured data set and the sample unstructured data set, thereby further improving the classification efficiency of structured data and unstructured data, and improving the efficiency of constructing the institution evaluation model.

[0110] For example, Figure 8 The input layer, data processing layer, vector representation layer, model layer, weight calculation layer and output layer are jointly trained, and the loss is constructed according to the difference between the sample evaluation score output by the output layer and the sample evaluation score label of the sample organization. The multiple structural layers are trained synchronously, and the combination of the various structural layers at the end of training is used as the organization evaluation model. In the application process, the unstructured data set and the structured data set of the organization to be evaluated can be directly input into the two corresponding input layers of the organization evaluation model for data processing, so that the final evaluation score of the organization evaluation model can be directly output through the output layer.

[0111] Exemplarily, a method for constructing an institutional evaluation model based on this system includes:

[0112] S1, obtain source data and the number of combination rules, and continue to S2;

[0113] S2. The source data is divided into an unstructured data set and a structured data set, and the unstructured data set and the structured data set are output, and the process continues to S3;

[0114] S3, perform data processing on the unstructured data set, the structured data set, and the historical institution ratings, output a floating-point vector, a feature vector, and a target vector, and continue to S4;

[0115] S3.1. Use the Sentence-BERT method to process unstructured datasets, map sentences and paragraphs into a high-dimensional dense vector space, and output floating-point vectors.

[0116] S3.2. Normalize and one-hot encode the structured dataset and historical institution scores, and output feature vectors and target vectors.

[0117] S4, determining whether the standard data set complies with the model input rules. If not, return to S3; if yes, continue to S5;

[0118] S5. Train the DNN algorithm based on the floating-point vector and the target vector to generate a DL_model evaluation model (unstructured data evaluation model). Train the linear algorithm based on the feature vector and the target vector to generate an ML_model evaluation model (structured data evaluation model). Output the DL_model and ML_model, and continue to S6.

[0119] S6. Output a combination weight set according to a preset number of combination rules;

[0120] S7. Traverse the combined weight set, adjust the weight ratio of DL_model and ML_model, and calculate the combined evaluation index value of the model;

[0121] S8. Select the combination model with the lowest combination evaluation index value, and output the evaluation model and combination weight value.

[0122] The vector representation layer outputs a standard data set, and the results include floating-point vectors, feature vectors, and target vectors corresponding to sample evaluation score labels.

[0123] Specifically, a floating-point vector T can be generated for each piece of text in the sample source data. i , the calculation formula is as follows:

[0124]

[0125] The floating-point vectors of all data are combined into a floating-point vector set T as follows:

[0126] T=[T1 T2 … T m ]

[0127] Among them, i is the long text information of the i-th data, v and n are the sizes of the floating-point vector space of the long text;

[0128] Specifically, after removing the attributes with high correlation, the remaining attributes are used as features to construct the feature dimension F as follows:

[0129]

[0130] Among them, d is the number of features; m is the number of data items; f ij is the jth dimension in the feature dimension and the i-th data.

[0131] Specifically, the historical institution ratings (sample evaluation score labels) constitute the target dimension S as follows:

[0132]

[0133] Among them, s i Score each piece of data historically.

[0134] Specifically, the training evaluation model is trained based on the floating-point vector set and the target vector training algorithm, traversing T and continuously adjusting its weight parameter w [1] 、w [2] and bias parameter b [1] , where the calculation formulas for each parameter are as follows:

[0135]

[0136] Among them, w [1] 、w [2] To train the weight parameters of different network layers in the unstructured model, network layers can be added according to the actual situation, that is, the weight parameter space is increased, b [1] is the bias parameter, S [1i] is the evaluation score of the i-th text through the model, such as Figure 9 As shown, Figure 9 is a first sample score S [1] Schematic diagram of the calculation method.

[0137] Specifically, the training algorithm trains the evaluation model based on the feature dimension vector and the target vector, and the calculation formula of the parameters involved is as follows:

[0138]

[0139] F.w [3] +b [2] =S [2]

[0140] Among them, w [3] is the weight parameter for training the structured model, b [2] is the bias parameter, S [2] is the second sample score

[0141] The final score is S=k1S [1] +k2S [2] , that is, output = k1DL_model + k2ML_model, and k1+k2=1.

[0142] It can be seen from the technical solutions provided in the above embodiments of this specification that the embodiments of this specification obtain the sample organization basic elements, sample scientific research capability elements and sample organization carrying capacity elements of the sample organization as sample source data; the sample source data is marked with the sample evaluation score label of the sample organization; according to the sample organization basic elements and sample scientific research capability elements, sample numerical data and sample character data are extracted as sample structured data sets; the sample long texts in the sample scientific research capability elements and sample organization carrying capacity elements are extracted as sample unstructured data sets; thereby dividing the sample source data into two major categories of data; converting the sample unstructured data set into a sample floating-point vector, and converting the sample structured data set into a sample feature vector; converting the sample floating-point vector and the sample evaluation score label An unstructured data evaluation model is input for score prediction to obtain a first sample score; and the sample feature vector and the sample evaluation score label are input for score prediction to obtain a second sample score; thereby, evaluation scores of the sample institutions obtained by prediction of the two types of data are obtained respectively; the combined weight set of the model is traversed again, and a target weight ratio is screened out from multiple weight ratios according to each weight ratio in the combined weight set, the first sample score, the second sample score and the sample evaluation score label; an institution evaluation model is constructed according to the unstructured data evaluation model, the structured data evaluation model and the target weight ratio; the institution evaluation model constructed using the method of the present invention has a higher accuracy rate, thereby being able to quickly and accurately predict the evaluation score of the institution to be evaluated.

[0143] The embodiment of this specification also provides a device for constructing an organization evaluation model, such as Figure 10 As shown, the device includes:

[0144] The sample data acquisition module 1010 is used to acquire the sample institution basic elements, sample scientific research capability elements, and sample institution carrying capacity elements of the sample institution as sample source data; the sample source data is annotated with the sample evaluation score label of the sample institution;

[0145] The sample structured data determination module 1020 is configured to extract sample numerical data and sample character data as a sample structured data set based on the sample organization basic elements and the sample scientific research capability elements;

[0146] A sample unstructured data determination module 1030 is configured to extract the sample long text from the sample scientific research capability element and the sample institutional carrying capacity element as a sample unstructured data set;

[0147] A sample vector conversion module 1040 is configured to convert the sample unstructured data set into a sample floating-point vector, and convert the sample structured data set into a sample feature vector;

[0148] The sample score prediction module 1050 is configured to input the sample floating-point number vector and the sample evaluation score label into the unstructured data evaluation model for score prediction to obtain a first sample score; and input the sample feature vector and the sample evaluation score label into the structured data evaluation model for score prediction to obtain a second sample score;

[0149] a target weight ratio determination module 1060 for traversing the combined weight set of the model and screening out a target weight ratio from a plurality of weight ratios based on each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label;

[0150] The evaluation model construction module 1070 is used to construct an institutional evaluation model based on the unstructured data evaluation model, the structured data evaluation model and the target weight ratio.

[0151] In some embodiments, the target weight ratio determination module includes:

[0152] an evaluation index value determining unit, configured to traverse the combined weight set of the model, and determine the combined evaluation index value of the model corresponding to each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label;

[0153] The target weight ratio determination unit is used to determine the optimal weight ratio of the combined evaluation index value that meets the preset conditions as the target weight ratio.

[0154] In some embodiments, the evaluation index value determination unit includes:

[0155] A weight ratio adjustment subunit, configured to input the combined weight set of the model into the weight calculation layer, traverse the combined weight set, and adjust the weight ratio of the unstructured data evaluation model and the structured data evaluation model;

[0156] a sample score determination subunit, configured to determine, based on the first sample score, the second sample score, and each weight ratio, a sample prediction evaluation score for the sample institution at each weight ratio;

[0157] The evaluation index value determination subunit is used to determine the combined evaluation index value of the model corresponding to each weight ratio according to the difference between the sample prediction evaluation score under each weight ratio and the sample evaluation score label.

[0158] In some embodiments, the sample vector conversion module includes:

[0159] A first data processing unit is configured to input the sample unstructured data set into an unstructured data processing layer for data cleaning to obtain a first sample standard data set;

[0160] a sample floating-point number vector extraction unit, configured to input the first sample standard data set into a first vector representation layer for text semantic extraction to obtain the sample floating-point number vector;

[0161] A second data processing unit is configured to input the sample structured data set into a structured data processing layer for data cleaning to obtain a second sample standard data set;

[0162] The sample feature vector determining unit is configured to input the second sample standard data set into the second vector representation layer to perform data correlation analysis to obtain the sample feature vector.

[0163] In some embodiments, the apparatus further comprises:

[0164] A first score prediction module is configured to input the sample floating-point number vector into an unstructured data evaluation network to perform evaluation score prediction to obtain a first prediction score;

[0165] a first loss determination module, configured to determine first loss data based on a difference between the first prediction score and the sample evaluation score label;

[0166] A first training module is used to adjust the parameters of the unstructured data processing layer, the first vector representation layer and the unstructured data evaluation network according to the first loss data until the training end condition is met, and determine the unstructured data evaluation network at the end of training as the unstructured data evaluation model.

[0167] In some embodiments, the training method of the structured data evaluation model includes:

[0168] A second score prediction module is used to input the sample feature vector into the structured data evaluation network to predict the evaluation score and obtain a second prediction score;

[0169] A second loss determination module, configured to determine second loss data based on a difference between the second prediction score and the sample evaluation score label;

[0170] The second training module is used to adjust the parameters of the structured data processing layer, the second vector representation layer and the structured data evaluation network according to the second loss data until the training end condition is met, and determine the structured data evaluation network at the end of training as the structured data evaluation model.

[0171] In some embodiments, the sample structured data determination module includes:

[0172] A first sample data extraction unit is configured to extract the number of projects, the number of tasks for each project, the budget for each project, and the number of people at each level in the organization team from the basic elements of the sample organization to obtain first sample structured data;

[0173] A second sample data extraction unit is used to extract the number of scientific research achievements in the sample scientific research capability element to obtain second sample structured data;

[0174] The sample structured data set construction unit is configured to construct the sample structured data set according to the first sample structured data and the second sample structured data.

[0175] In some embodiments, the apparatus further comprises:

[0176] The module for acquiring source data to be evaluated is used to acquire the basic elements, scientific research capability elements and institutional carrying capacity elements of the institution to be evaluated as the source data to be evaluated;

[0177] An unstructured data set determination module is used to extract a structured data set based on the institutional basic elements and the scientific research capability elements; and extract the long text in the scientific research capability elements and the institutional carrying capacity elements as an unstructured data set;

[0178] A feature vector conversion module, configured to convert the unstructured data set into a floating point vector and convert the structured data set into a feature vector;

[0179] A first evaluation score prediction module, configured to input the floating-point number vector into the unstructured data evaluation model for score prediction to obtain a first evaluation score;

[0180] A second evaluation score prediction module, configured to input the feature vector into the structured data evaluation model for score prediction to obtain a second evaluation score;

[0181] An evaluation score determination module is used to obtain the evaluation score of the organization to be evaluated based on the first evaluation score, the second evaluation score and the target weight ratio.

[0182] In some embodiments, the evaluation score determination module includes:

[0183] a weight determination unit, configured to determine a first weight of the unstructured data evaluation model and a second weight of the structured data evaluation model according to the target weight ratio;

[0184] a product calculation unit, configured to calculate a first product of the first evaluation score and the first weight, and a second product of the second evaluation score and the second weight;

[0185] An evaluation score calculation unit is used to calculate the sum of the first product and the second product to obtain the evaluation score of the institution to be evaluated.

[0186] The device and method embodiments in the device embodiments are based on the same inventive concept.

[0187] In the embodiments of the present invention, the term "module" or "unit" refers to a computer program or a part of a computer program that has a predetermined function and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories), or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0188] An embodiment of this specification provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the method for constructing an institutional evaluation model provided in the above method embodiment.

[0189] An embodiment of the present invention also provides a computer storage medium, which can be set in a terminal to store at least one instruction or at least one program related to a method for constructing an institutional evaluation model in a method embodiment. The at least one instruction or at least one program is loaded and executed by the processor to implement the method for constructing an institutional evaluation model provided in the above method embodiment.

[0190] Embodiments of the present invention also provide a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to implement the method for constructing an institutional evaluation model provided in the above-described method embodiment.

[0191] Optionally, in the embodiments of this specification, the storage medium may be located in at least one of the multiple network servers of the computer network. Optionally, in this embodiment, the storage medium may include, but is not limited to, various media that can store program code, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0192] The memory described in the embodiments of this specification can be used to store software programs and modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory may mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, application programs required for functions, etc.; the data storage area can store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory may also include a memory controller to provide the processor with access to the memory.

[0193] The method for constructing the institutional evaluation model provided in the embodiments of this specification can be executed in a mobile terminal, a computer terminal, a server or a similar computing device. Taking running on a server as an example, Figure 11 This is a hardware structure diagram of a server for a method of constructing an institution evaluation model provided in an embodiment of this specification. Figure 11As shown, the server 1100 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPUs) 1110 (the central processing unit 1110 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 1130 for storing data, and one or more storage media 1120 (such as one or more mass storage devices) for storing application programs 1123 or data 1122. Among them, the memory 1130 and the storage medium 1120 can be temporary storage or permanent storage. The program stored in the storage medium 1120 may include one or more modules, each module may include a series of instruction operations on the server. Furthermore, the central processing unit 1110 may be configured to communicate with the storage medium 1120 to execute a series of instruction operations in the storage medium 1120 on the server 1100. The server 1100 may also include one or more power supplies 1160, one or more wired or wireless network interfaces 1150, one or more input and output interfaces 1140, and / or one or more operating systems 1121, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0194] The input / output interface 1140 can be used to receive or send data via a network. Specific examples of the aforementioned network may include a wireless network provided by the communication provider of the server 1100. In one embodiment, the input / output interface 1140 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, the input / output interface 1140 can be a radio frequency (RF) module for wireless communication with the Internet.

[0195] It can be understood by those skilled in the art that Figure 11 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 11 More or fewer components than shown, or with Figure 11 Different configurations shown.

[0196] It can be seen from the embodiments of the method, device, equipment or storage medium for constructing the institution evaluation model provided by the above-mentioned present invention that the present invention obtains the sample institution basic elements, sample scientific research capability elements and sample institution carrying capacity elements of the sample institution as sample source data; the sample source data is marked with the sample evaluation score label of the sample institution; according to the sample institution basic elements and sample scientific research capability elements, sample numerical data and sample character data are extracted as sample structured data sets; the sample long texts in the sample scientific research capability elements and sample institution carrying capacity elements are extracted as sample unstructured data sets; thereby dividing the sample source data into two major categories of data; converting the sample unstructured data set into a sample floating-point vector, and converting the sample structured data set into a sample feature vector; converting the sample floating-point vector and the sample This evaluation score label is input into the unstructured data evaluation model for score prediction to obtain a first sample score; and the sample feature vector and the sample evaluation score label are input into the structured data evaluation model for score prediction to obtain a second sample score; thereby, the evaluation scores of the sample institutions obtained by predicting the two types of data are obtained respectively; then the combined weight set of the model is traversed, and the target weight ratio is screened out from multiple weight ratios according to each weight ratio in the combined weight set, the first sample score, the second sample score and the sample evaluation score label; an institution evaluation model is constructed according to the unstructured data evaluation model, the structured data evaluation model and the target weight ratio; the institution evaluation model constructed using the method of the present invention has a higher accuracy rate, thereby being able to quickly and accurately predict the evaluation score of the institution to be evaluated.

[0197] It should be noted that the order in which the embodiments of this specification are presented is for illustrative purposes only and does not represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions are of specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in an order different from that in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0198] The various embodiments in this specification are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the device, equipment, and storage medium embodiments are generally similar to the method embodiments, so their descriptions are relatively simplified. For relevant portions, refer to the descriptions of the method embodiments.

[0199] Those skilled in the art will understand that all or part of the steps of implementing the above embodiments may be accomplished by hardware, or by a program instructing the relevant hardware to accomplish the steps. The program may be stored in a computer storage medium, and the above-mentioned storage medium may be a read-only memory, a disk, or an optical disk, etc.

[0200] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for constructing an institution evaluation model, characterized in that: The method comprises: Obtaining sample institution basic elements, sample scientific research capability elements, and sample institution carrying capacity elements of the sample institution as sample source data; the sample source data is annotated with a sample evaluation score label of the sample institution; Extracting sample numerical data and sample character data as a sample structured data set based on the sample organization basic elements and sample scientific research capability elements; Extracting sample long texts from the sample scientific research capability elements and the sample institutional carrying capacity elements as sample unstructured data sets; Converting the sample unstructured data set into a sample floating-point vector, and converting the sample structured data set into a sample feature vector; Inputting the sample floating-point number vector and the sample evaluation score label into the unstructured data evaluation model for score prediction to obtain a first sample score; and inputting the sample feature vector and the sample evaluation score label into the structured data evaluation model for score prediction to obtain a second sample score; Traversing the combined weight set of the model, and screening a target weight ratio from multiple weight ratios according to each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label; An institutional evaluation model is constructed based on the unstructured data evaluation model, the structured data evaluation model, and the target weight ratio.

2. The method according to claim 1, characterized in that The combined weight set of the ergodic model is traversed, and a target weight ratio is screened out from a plurality of weight ratios according to each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label, including: Traversing the combined weight set of the model, and determining the combined evaluation index value of the model corresponding to each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label; The optimal weight ratio of the combined evaluation index value that meets the preset conditions is determined as the target weight ratio.

3. The method according to claim 2, characterized in that The traversing the combined weight set of the model, and determining the combined evaluation index value of the model corresponding to each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label, includes: Inputting the combined weight set of the model into the weight calculation layer, traversing the combined weight set, and adjusting the weight ratio of the unstructured data evaluation model and the structured data evaluation model; Determining a sample prediction evaluation score of the sample institution at each weight ratio based on the first sample score, the second sample score, and each weight ratio; According to the difference between the sample prediction evaluation score under each weight ratio and the sample evaluation score label, the combined evaluation index value of the model corresponding to each weight ratio is determined.

4. The method according to claim 1, wherein The converting the sample unstructured data set into a sample floating-point vector, and converting the sample structured data set into a sample feature vector, comprises: Inputting the sample unstructured data set into the unstructured data processing layer for data cleaning processing to obtain a first sample standard data set; Inputting the first sample standard data set into a first vector representation layer to perform text semantic extraction to obtain the sample floating-point number vector; Inputting the sample structured data set into the structured data processing layer for data cleaning processing to obtain a second sample standard data set; The second sample standard data set is input into the second vector representation layer to perform data correlation analysis to obtain the sample feature vector.

5. The method according to claim 4, characterized in that The training method of the unstructured data evaluation model includes: Inputting the sample floating-point number vector into an unstructured data evaluation network to perform evaluation score prediction to obtain a first prediction score; Determining first loss data based on a difference between the first prediction score and the sample evaluation score label; Adjust the parameters of the unstructured data processing layer, the first vector representation layer, and the unstructured data evaluation network according to the first loss data until the training end condition is met, and determine the unstructured data evaluation network at the end of training as the unstructured data evaluation model.

6. The method according to claim 4, characterized in that The training method of the structured data evaluation model includes: Inputting the sample feature vector into a structured data evaluation network to predict an evaluation score to obtain a second prediction score; Determining second loss data based on a difference between the second prediction score and the sample evaluation score label; The parameters of the structured data processing layer, the second vector representation layer and the structured data evaluation network are adjusted according to the second loss data until the training end condition is met, and the structured data evaluation network at the end of training is determined as the structured data evaluation model.

7. The method according to claim 1, characterized in that The extracting of sample numerical data and sample character data as a sample structured data set based on the sample organization basic elements and the sample scientific research capability elements includes: Extracting the number of projects, the number of tasks for each project, the budget for each project, and the number of people at each level in the organization team from the basic elements of the sample organization to obtain first sample structured data; Extracting the number of scientific research achievements in the sample scientific research capability element to obtain second sample structured data; The sample structured data set is constructed according to the first sample structured data and the second sample structured data.

8. The method according to claim 1, characterized in that After constructing the organization evaluation model based on the unstructured data evaluation model, the structured data evaluation model, and the target weight ratio, the method includes: Obtain the basic elements, scientific research capability elements, and institutional carrying capacity elements of the institution to be evaluated as source data to be evaluated; Extracting a structured data set based on the institutional basic elements and scientific research capability elements; and extracting long texts from the scientific research capability elements and institutional carrying capacity elements as an unstructured data set; Converting the unstructured data set into a floating point vector and converting the structured data set into a feature vector; Inputting the floating-point number vector into the unstructured data evaluation model for score prediction to obtain a first evaluation score; Inputting the feature vector into the structured data evaluation model to perform score prediction to obtain a second evaluation score; An evaluation score of the institution to be evaluated is obtained according to the first evaluation score, the second evaluation score and the target weight ratio.

9. The method according to claim 8, characterized in that Obtaining the evaluation score of the institution to be evaluated according to the first evaluation score, the second evaluation score, and the target weight ratio includes: Determining a first weight of the unstructured data evaluation model and a second weight of the structured data evaluation model according to the target weight ratio; Calculating a first product of the first evaluation score and the first weight, and a second product of the second evaluation score and the second weight; The sum of the first product and the second product is calculated to obtain an evaluation score of the institution to be evaluated.

10. A device for constructing an organization evaluation model, characterized in that: The device comprises: A sample data acquisition module is used to acquire sample institution basic elements, sample scientific research capability elements, and sample institution carrying capacity elements of a sample institution as sample source data; the sample source data is annotated with a sample evaluation score label of the sample institution; A sample structured data determination module is used to extract sample numerical data and sample character data as a sample structured data set based on the sample organization basic elements and the sample scientific research capability elements; A sample unstructured data determination module is used to extract the sample long text from the sample scientific research capability element and the sample institutional carrying capacity element as a sample unstructured data set; A sample vector conversion module, configured to convert the sample unstructured data set into a sample floating-point vector, and convert the sample structured data set into a sample feature vector; A sample score prediction module is configured to input the sample floating-point number vector and the sample evaluation score label into an unstructured data evaluation model for score prediction to obtain a first sample score; and input the sample feature vector and the sample evaluation score label into a structured data evaluation model for score prediction to obtain a second sample score; a target weight ratio determination module, configured to traverse the combined weight set of the model and select a target weight ratio from a plurality of weight ratios based on each weight ratio in the combined weight set, the first sample score, the second sample score, and the sample evaluation score label; An evaluation model construction module is used to construct an institutional evaluation model based on the unstructured data evaluation model, the structured data evaluation model and the target weight ratio.

Images